SILVERBERG, MICHAEL J. MPH*; AHDIEH, LINDA MPH, PhD*; MUÑOZ, ALVARO PhD*; ANASTOS, KATHRYN MD†; BURK, ROBERT D. MD‡; CU-UVIN, SUSAN MD§; DUERR, ANN MD, PhD∥; GREENBLATT, RUTH M. MD¶; KLEIN, ROBERT S. MD†‡; MASSAD, STEWART MD#; MINKOFF, HOWARD MD**; MUDERSPACH, LAILA MD††; PALEFSKY, JOEL MD¶; PIESSENS, EVA MD‡‡; SCHUMAN, PAULA MD§§; WATTS, HEATHER MD∥∥; SHAH, KEERTI V. MD, DrPH*
IN THE UNITED STATES, the most common sexually transmitted infection is that by human papillomaviruses (HPVs). 1 The most serious consequences of genital HPV infections are high-grade squamous intraepithelial lesions, which can progress to invasive cervical cancer, the third most common cancer among women worldwide after breast and colorectal cancer and the second most common cancer in developing countries after breast cancer. 2 Few investigations have focused on genital warts, one of the most frequent clinical manifestations of HPV infection. An estimated 1% of the sexually active US population have prevalent genital warts. 3 These are benign lesions that may be located anywhere in the genital tract but are most frequent at the external genitalia. 4
The nononcogenic HPV types 6 and 11 are responsible for the large majority of genital warts. These two types are recovered from 70% to 100% of genital wart tissues 5–10 but almost never from cervical cancer or other genital malignancies. HPV 6 and 11 are also responsible for wartlike lesions in the oral cavity and on the conjunctiva. 11 In addition, a maternal history of genital warts is an established risk factor for the occurrence of juvenile onset recurrent respiratory papillomatosis (JORRP); between 50% and 70% of JORRP cases involve a history of maternal genital warts during pregnancy or at the time of delivery. 12,13
As compared with HIV-seronegative individuals, HIV-seropositive individuals have a higher prevalence and longer duration of genital tract HPV infections. 14–20 HIV-seropositive individuals are also more likely to be infected with multiple HPV types. 10,21 Thus, the risk of genital wart development is increased in HIV-seropositive women and may be the result of HIV-associated immunologic abnormalities. 22,23 In the current study, we compared HPV 6 and 11 infections and their relationship to genital warts in HIV-infected and HIV-negative women in two large cohorts and examined the role of HIV-associated immunosuppression on the course of HPV infection.
Study Population and Design
The Women's Interagency HIV Study (WIHS) is a large cohort consisting of 569 HIV-seronegative and 2059 HIV-seropositive women followed every 6 months. 24 Participants were enrolled at six sites between October 1994 and November 1995 (Bronx/Manhattan, NY; Brooklyn, NY; Chicago, IL; Los Angeles, CA; San Francisco, CA; and Washington, DC). They were recruited from a variety of sources, including HIV primary care clinics, hospital-based programs, research programs, community outreach sites, women's support groups, drug rehabilitation programs, HIV testing sites, and referrals from enrolled participants. The HIV-seropositive and -seronegative cohorts were frequency-matched on age, race/ethnicity, level of education, injection drug use since 1978, and total number of sex partners since 1980.
The HIV Epidemiology Research Study (HERS), with sites in Baltimore, MD, the Bronx, NY, Detroit, MI, and Providence, RI, has been described in detail. 25 Between April 1993 and January 1995, 439 HIV-seronegative and 871 HIV-seropositive women, also followed at 6-month intervals, were enrolled. Participants were recruited via community gathering places, community newspapers, infectious disease and drug treatment clinics, previous clinical studies, and word of mouth. HIV-seronegative individuals were recruited over the same time period using similar sources. Report of injection drug use since 1985 and sexual risk behaviors were monitored during recruitment to ensure comparability between the HIV-seropositive and -seronegative cohorts. In contrast to the WIHS, potential HERS participants were excluded if they reported a previously diagnosed AIDS-defining illness (according to the 1987 AIDS surveillance definition of the Centers for Disease Control and Prevention [Atlanta]).
For both studies, the protocols were reviewed and approved by the institutional review boards of the participating institutions, and the women in the studies provided written informed consent.
At baseline and at each 6-month follow-up visit, participants in both cohorts completed interviewer-administered questionnaires to assess sociodemographics; medical/health history; obstetric, gynecologic, and contraceptive history; tobacco, alcohol, and drug use history; sexual behaviors; health care access; and psychosocial measures. Medical charts were abstracted to identify information on hospitalizations and HIV-related illnesses.
Participants in both studies underwent a physical and gynecologic examination at each study visit, primarily aimed at identifying HIV-associated conditions. Investigative personnel in both studies received centralized training to ensure standardized data collection during the clinical examination and interviews. Annual site visits were performed to monitor adherence to study protocols. In the WIHS, vulvar, vaginal, cervical, and anal lesions, including genital warts, were identified by visual inspection and characterized in terms of number, location, size, depth, and morphology. In the HERS, genital warts were identified and classified as external, cervical, or vaginal warts. Clinically diagnosed genital warts were most likely condylomata acuminata, the most common morphologic type.
Papanicolaou smear and perianal examinations were performed on all participants. Referrals were given for colposcopy, biopsy, or treatment of dysplasia if indicated. Biologic specimens, including blood, urine, and a cervicovaginal lavage sample, were collected and stored. The procedure for the collection of cervicovaginal lavage specimens was the same in both cohorts and included the injection of 10 ml of sterile normal saline against the cervical os and the exocervix. The fluid was then taken up by a disposable pipette and transferred to a polypropylene tube.
HPV testing was performed on cervicovaginal lavage specimens in the laboratories of Drs. Burk (Albert Einstein College of Medicine) and Palefsky (University of California, San Francisco) for the WIHS and Dr. Shah (The Johns Hopkins School Hygiene and Public Health) for the HERS. Testing has been completed for 4 WIHS visits and 10 HERS visits. Details of the HPV testing have been described elsewhere. 14,26 In brief, the testing was performed using polymerase chain reaction with L1 consensus primers. 26 Amplification of β-globin DNA was used as a positive control for the presence of amplifiable DNA. Type-specific probes were used to identify individual HPV types, including 6, 11, 16, 18, 26, 31, 32, 33, 35, 39, 40, 45, 51, 52, 53, 54, 55, 56, 58, 59, 61, 66, 68, 69, 70, 73, 83, and 84. Samples positive for the consensus probe but negative for all type-specific probes were considered HPV-positive, untyped.
HIV status at baseline was determined by enzyme-linked immunosorbent assay (ELISA) and confirmatory Western blot. CD4+ T-lymphocyte levels (cells/μl) were determined at each study visit using standardized flow cytometry. 27
HPV infection was categorized as HPV 6, HPV 11, other HPV, or no HPV. Coinfections with HPV type 6 or 11 and other HPV types were categorized as HPV 6 or HPV 11 infections. The prevalence of HPV at baseline, defined as the enrollment visit, was determined according to HIV status, and the groups were compared using the Pearson chi-square statistic. Incident HPV 6 or 11 infections were defined as a positive HPV result during follow-up after a negative baseline HPV test for these types. Incidence of other HPV types was defined as detection of HPV types other than HPV 6 or 11 during follow-up after an initial negative baseline HPV test for other types. We calculated the rate of incident HPV infections by HIV status for each cohort, expressed as cases per 100 person-years. We recognize that “incident” cases may not be truly incident but rather reactivated latent infections. The Fisher exact test was used to compare the rates of HPV infection between HIV-seronegative and HPV-seropositive participants.
The prevalence and incidence of genital warts were also determined by HIV status for each cohort. Prevalent genital warts were defined as lesions observed at the clinical examination at the baseline visit. Incident genital warts were defined as lesions observed during follow-up among women without genital warts at baseline. The Fisher exact test was used to compare the rates between HIV-seronegative and HIV-seropositive participants.
Subsequent analyses defined visits rather than participants as the unit of analysis. Thus, for some women there were multiple observations. As before, the HPV infection categories were HPV 6 or 11 infection, other HPV infection, and no HPV infection. Individuals coinfected with HPV types 6 and 11 and other types were categorized as having HPV 6 or 11 infections. The presence of genital warts and CD4+ cell count were treated as time-dependent variables and determined at each study visit. The distribution of visits by HPV type, HIV status, and genital wart status was determined. We calculated the prevalence of genital warts by HIV status and level of immunodeficiency (CD4 cell count ≥501, 351–500, 201–350, or ≤200/μl). Within each HPV category, a Mantel-Haenszel test for trend was used to evaluate a possible increase in genital wart prevalence with decreasing CD4+ cell count.
Using statistical models, we quantified (1) the extent to which HPV infection increased the odds of prevalent genital warts and (2) the modification of this relationship by HIV infection and CD4+ cell count. Variables in the final model included two dummy variables to account for the three categories of HPV infection, a two-way interaction term between HIV and HPV infection, and a three-way interaction term between HIV, HPV, and CD4+ cell count. CD4+ cell count was dichotomized as ≥201 and ≤200 cells/μl for the final model. The effect of HIV infection and CD4+ cell count on genital warts in the absence of HPV infection was not examined in the statistical models (i.e., the main effects of HIV infection and CD4+ cell count were not included). The GENMOD procedure in SAS version 8.0 (SAS Institute, Cary, NC) was used with an unstructured correlation matrix type to fit statistical models, with use of a generalized estimating equation that accounted for correlated observations.
Despite the availability of data from the HERS through visit 10, the analysis was restricted through visit 6 to match the follow-up in the WIHS by calendar time. We analyzed the cohorts separately to ascertain the reproducibility of our observations, because each had a sufficient number of women to enable the study objectives to be addressed.
Our analyses incorporated data from 551 HIV-seronegative and 2032 HIV-seropositive women from the WIHS and 420 HIV-seronegative and 863 HIV-seropositive women from the HERS. Twenty-three women who were HIV-seronegative at baseline and seroconverted during follow-up were excluded. Missing HPV or clinical examination data resulted in the exclusion of 528 (6.3%) and 439 (6.8%) observations for the WIHS and HERS participants, respectively. Data from HIV-seropositive women whose CD4+ cell counts were missing (231 [2.9%] in the WIHS and 115 [1.9%] in the HERS) were excluded from relevant analyses.
Descriptive statistics are displayed in Table 1 by cohort and HIV status. HIV-seronegative and -seropositive women were similar in both cohorts in age, race, and injection drug use. In the HERS, HIV-seronegative women reported more education than did HIV-seropositive women. In the WIHS, HIV-seronegative and HIV-seropositive women accounted for 1636 and 6277 visits, respectively; in the HERS, these totals were 1931 and 4108. Women in the WIHS and HERS had a median of three and five observations, respectively. In a comparison of factors across studies, injection drug use in the 6 months before baseline was twice as likely among women in the HERS. The median CD4+ cell count among HIV-seropositive women was lower at baseline in the WIHS than in the HERS (329 versus 376 cells/μl;P < 0.001), which may be explained by the exclusion of women with AIDS in the HERS only.
Relationship of HIV to HPV and HIV to Genital Warts
HPV 6 or 11 DNA detection at baseline was markedly higher among HIV-seropositive women in both studies (Table 2; 5.0% versus 0.9% in the WIHS, 4.3% versus 1.2% in the HERS). Other HPV types were commonly detected at baseline among HIV-seronegative women (29.0% in the WIHS, 26.0% in the HERS) and at an even higher proportion in HIV-seropositive women (58.7% in the WIHS, 59.9% in the HERS). The difference in HPV detection at baseline between HIV-seronegative and HIV-seropositive women was significant (P < 0.001) for women infected with HPV type 6 or 11, as well as those infected with other types in both cohorts. HIV-seropositive women also had a strikingly higher baseline prevalence of genital warts than did HIV-seronegative women. Baseline rates were 9.8% versus 3.1% (P < 0.001) in the WIHS and 13.6% versus 5.0% (P < 0.001) in the HERS.
HIV infection was also associated with an increased incidence of HPV and genital warts in both cohorts. As presented in Table 3, the incidence (cases/100 person-years) of HPV 6 or 11 DNA after baseline among HIV-seronegative women was 1.0 and 0.6 in the WIHS and HERS, respectively; among HIV-seropositive women, the rates were 3.1 and 3.7. Differences in the incidence of HPV 6 or 11 DNA between HIV-seronegative and HIV-seropositive women were significant for both cohorts (P = 0.003 in the WIHS;P < 0.001 in the HERS). The incidence of other HPV types in the WIHS and HERS was 18.3 and 24.7 for HIV-seronegative women and 40.8 and 43.8 for HIV-seropositive women. Differences in the incidence of other HPV infections were also significant for both cohorts (P < 0.001 in the WIHS;P < 0.001 in the HERS). Detection rates of new genital warts were 2.2 and 8.9 (P < 0.001) in HIV-seronegative and HIV-seropositive women in the WIHS, respectively, and 5.8 and 14.3 (P < 0.001) in the HERS in these same groups.
HPV 6 was more commonly detected than HPV 11 in both HIV-seronegative and HIV-seropositive women and in both cohorts. Overall, HPV 6 was identified in 343 specimens and HPV 11 in 180 specimens, giving a ratio of approximately 2:1. Coinfection with HPV type 6 or 11 and other HPV types was common in both cohorts, particularly among HIV-seropositive women. Overall, among HPV 6 or 11 infections, 43% were coinfections with other HPV types among HIV-seronegative women, and 75% were coinfections with other HPV types among HIV-seropositive women.
Relationship Between HPV, HIV, and Genital Warts
The distribution of HPV types by HIV and genital wart status is presented in Table 4. The prevalence of detectable HPV 6 or 11 in the cervicovaginal lavage was low among women with genital warts, despite the established causal link between infection with these HPV types and genital warts. Among HIV-seronegative women, HPV 6 or 11 was detected at 7.3% and 9.3% of genital wart–positive visits in the WIHS and HERS, respectively. Among HIV-seropositive women, HPV 6 or 11 was detected in 18.5% and 18.8% of genital wart–positive visits in the WIHS and HERS, respectively. Nevertheless, the presence of genital warts was associated with an increased prevalence of HPV, particularly HPV 6 or 11, in comparison with cases involving no genital warts. Among HIV-seronegative women, the presence of genital warts was associated with increases of 12.2-fold and 23.3-fold in the prevalence of HPV 6 or 11 in the WIHS and HERS, respectively. The relative increase in prevalence of HPV 6 or 11 among HIV-seropositive women was 6.6- and 6.5-fold in the WIHS and HERS, respectively. Other HPV types were also more frequent in women with genital warts than in those without genital warts, regardless of HIV status, in the WIHS and HERS, although the differences were not as pronounced.
The prevalence of genital warts, as related to HPV infection and HIV immunodeficiency, at follow-up visits in the WIHS and HERS is presented in Figure 1. In both studies, the percentage of visits by women with genital warts increased with decreasing CD4+ cell counts for all categories of HPV infection, particularly for HPV 6 or 11 infection (P for trend = 0.008 [WIHS];P for trend = 0.004 [HERS]). The increase in genital warts with other HPV types was modest but statistically significant (P for trend = 0.038 [WIHS];P for trend <0.001 [HERS]). Of note, HIV-seronegative women with HPV 6 or 11 infection in the HERS had genital warts more frequently (8 of 15 visits) than HIV-seropositive women with ≥501 CD4 cells/μl (4 of 22 visits), which does not conform to the overall pattern. This discrepancy is mainly attributable to one woman with genital warts for whom there were five observations in the HIV-seronegative group. The highest probability of genital warts was for visits of women with HPV 6 or 11 infection, HIV infection, and CD4+ cell counts <200/μl: 46.3% in the WIHS and 60.2% in the HERS. As described in Methods, subsequent analyses accounted for the inherent correlation of repeated observations for an individual.
Data on the joint contributions of HPV type, HIV infection, and CD4+ cell count on the prevalence of genital warts are presented in Table 5 and Figure 2. In the WIHS, the odds of having genital warts at HPV 6–positive or HPV 11–positive visits (relative to observations with no HPV) increased from 5.1 (95% CI: 2.9–8.8) among HIV-seronegative women to 8.8 (95% CI: 6.1–12.8) among HIV-seropositive women with ≥201 CD4 cells/μl and to 12.8 (95% CI: 8.8–18.8) among HIV-seropositive women with ≤200 CD4 cells/μl. This trend was also observed for other HPV types in these same groups: 1.6 (95% CI: 1.0–2.6), 2.8 (95% CI: 2.2–3.6), and 4.1 (95% CI: 3.1–5.4). Similar associations were found in the HERS but were of lower magnitudes. For HPV 6 or 11 infection, the odds were 2.7 (95% CI: 1.6–4.6) among HIV-seronegative women, 4.9 (95% CI: 3.2–7.7) among HIV-seropositive women with ≥201 CD4 cells/μl, and 5.3 (95% CI: 3.3–8.5) among HIV-seropositive women with ≤200 CD4 cells/μl. For other types, the odds were 1.0 (95% CI: 0.7–1.4), 1.8 (95% CI: 1.5–2.3), and 2.0 (95% CI: 1.4–2.6) in these same groups. The interaction by HIV status was significant for the WIHS (P = 0.025) and HERS (P = 0.015), whereas the interaction by CD4+ cell count was significant in the WIHS (P = 0.001) but not in the HERS (P = 0.61).
Recent investigations of the natural history of HPV have focused on types most closely associated with cervical cancer, the third most common cancer among women worldwide and the second most common cancer in developing countries. 2 HPV types 6 and 11 are rarely found in cervical cancer cases and thus have received less attention. These virus types, however, are primarily responsible for genital warts, one of the most prevalent sexually transmitted diseases worldwide. In addition, genital warts are an established risk factor for the occurrence of JORRP. 12,13
In the current study, we had the opportunity to investigate these infections in two large independent cohorts of HIV-seronegative and -seropositive women. These studies had similar protocols and examined women with similar behavior profiles. The results of the two studies were virtually identical in most respects. HIV seropositivity was associated with an increased prevalence and incidence of both HPV DNA and genital warts. The HPV rates were of similar magnitude in the two studies, but the HERS revealed higher rates of genital warts. Next, we quantified the increased odds of genital warts in HIV-seropositive women, particularly among those with advanced HIV disease (i.e., with a CD4+ cell count ≤200/μl).
The increased genital wart prevalence with higher levels of immunodeficiency was most pronounced among women infected with HPV 6 or 11, although a small increase was also detected in women infected with other HPV types. In both studies, the relative effect of HIV and associated immunodeficiency on genital warts was pronounced in women infected with HPV types 6 and 11, although, of interest, relatively few cases of genital warts were associated with the shedding of these viruses in the cervicovaginal lavage. The fact that the prevalence of genital warts was higher than the prevalence of HPV type 6 or 11 in the cervicovaginal lavage suggests that genital warts are localized lesions.
Previous studies 17,19,20 have demonstrated that HIV infection markedly increases the duration of HPV infection, which may be the most important reason for the higher prevalence among HIV-seropositive women. In a recent study, prevalent infections resolved at the following rates: for 50% of HIV-seronegative women, in approximately 6 months; for 50% of HIV-seropositive women, in approximately 24 months; and for 50% of HIV-seropositive women with low CD4+ cell counts (<200/μl), in >36 months. 28
HIV infection may also enhance the detectability of HPV because HIV-seropositive women may have greater amounts of virus than HIV-seronegative women. 29 Increased susceptibility of HIV-seropositive women to HPV infection is an additional possibility, but this has not been demonstrated in previous studies. The higher prevalence in HIV-seropositive women is unlikely to be due to increased high-risk behavior among these women. The HIV-seronegative and HIV-seropositive cohorts were matched on risk behaviors in both studies, and, furthermore, previous studies 23,30 have shown that high-risk sexual behaviors may decrease with increasing infirmity in HIV-seropositive individuals.
Any factor that increases HPV prevalence or incidence would also increase the prevalence of genital warts. The increased prevalence of genital warts with increased immunosuppression is most likely a result of the reduced ability of the immune system to control HPV infections. 21 Reactivation of latent HPV infections may be enhanced in HIV-seropositive individuals. Thus, we could not distinguish whether our observations were new infections or recurrent prevalent infections. Another explanation for our observed results is that HIV infection increases the number of HPV types with which an individual is coinfected, which may increase the probability of developing genital warts.
Brown et al. 10 demonstrated that several HPV types were detected from the genital wart lesions of only immunosuppressed patients. This suggests a causal relationship for HPV types other than types 6 and 11. Alternatively, women with other HPV types in the cervicovaginal lavage may also have a higher prevalence of genital warts as a result of an increased probability of unrecognized local HPV 6 or 11 infections. However, we found that infection with other HPV types, in addition to HPV 6 and 11, in the cervicovaginal lavage did not increase the probability of genital warts in comparison with HPV 6 or 11 infection alone (data not shown), despite higher rates of coinfection in HIV-seropositive women. These possibilities should be further examined.
The main difference between the results of the two studies was that the relationship between CD4+ cell count and the increase in genital warts was more pronounced in the WIHS than in the HERS (Table 5 and Figure 2). The main reason for this finding was that the prevalence of genital warts in the reference group, i.e., HPV-negative women, was greater in the HERS than in the WIHS. The rates were 3.3% (HERS) and 1.5% (WIHS) in HIV-seronegative women and 8.3% (HERS) and 3.1% (WIHS) in HIV-seropositive women. The reasons for these differences are not known. Differences in HIV antiretroviral experience between participants in the two studies may also explain the decreased magnitude of effect in HERS. According to a recent article by Heard et al., 31 triple-combination antiretroviral therapy for HIV disease may help to produce regression of HPV-associated lesions. We did not examine this possibility in our current study.
Some limitations of this study should be addressed. First, less advanced genital warts may have been overlooked. Such misclassification would have diluted the observed association. In addition, we did not evaluate the role of genital wart treatments. However, effective therapies would be expected to only reduce the observed prevalence of genital warts at subsequent visits and would not bias relative effect measures. Finally, the number of women infected with HPV type 6 or 11, particularly HIV-seronegative women, was small. Nevertheless, the trends we reported are clear and were replicated in two studies.
In summary, we found a strong correlation between increasing immunodeficiency and genital wart prevalence in two large prospective studies, particularly in women infected with HPV type 6 or 11. Strategies to increase the cellular immune response may have important implications for HPV disease progression. As such, studies examining the effect of antiretroviral therapy on HPV disease are warranted, because the reconstituted immune system may be more capable of resolving genital warts.
The most effective method for controlling HPV-associated disease would be to prevent initial infection, and efforts to develop effective prophylactic vaccines for the prevention of HPV infection are ongoing. The findings presented in this report highlight the need for vaccines to induce protection against HPV types 6 and 11, the causative agents of genital warts, in addition to the HPV types associated with cervical cancer. The efficacy of HPV vaccines in HIV-seropositive individuals, including those with HIV-induced immunosuppression, should also be considered because of the increased probability of clinical disease in such persons.
The HIV Epidemiology Research Study (HERS) group includes: Robert S. Klein, MD, Ellie Schoenbaum, MD, Julia Arnsten, MD, MPH, Robert D. Burk, MD, Chee Jen Chang, PhD, Penelope Demas, PhD, and Andrea Howard, MD, MSc, from Montefiore Medical Center and the Albert Einstein College of Medicine; Paula Schuman, MD, and Jack Sobel, MD, from the Wayne State University School of Medicine; Anne Rompalo, MD, David Vlahov, PhD, and David Celentano, PhD, from The Johns Hopkins University School of Medicine; Charles Carpenter, MD, Kenneth Mayer, MD, Susan Cu-Uvin, MD, Timothy Flanigan, MD, Joseph Hogan, ScD, Valerie Stone, MD, Karen Tashima, MD, and Josiah Rich, MD, from the Brown University School of Medicine; Ann Duerr, MD, PhD, Lytt I. Gardner, PhD, Scott D. Holmberg, MD, Denise J. Jamieson, MD, MPH, Janet S. Moore, PhD, Ruby M. Phelps, BS, Dawn K. Smith, MD, MPH, and Dora Warren, PhD, from the Centers for Disease Control and Prevention; and Katherine Davenny, MPH, from the National Institute of Drug Abuse.
The Women's Interagency HIV Study (WIHS) Collaborative Study Group members are as follows. New York City/Bronx Consortium: Bronx-Lebanon Hospital/Montefiore Medical Center (Kathryn Anastos, MD, Ed Telzak, MD, Ann Danoff, MD, Jessica Justman, MD, Esther Robison, PhD); Beth Israel Hospital (Usha Mathur-Wagh, MD); Mt. Sinai Medical Center (Henry S. Sacks, MD, PhD, Alejandra Gurtman, MD); Wadsworth Laboratories (Barbara Weiser, MD, Harold Burger, MD); Community Advisor (Amirah Waleed). Brooklyn, NY: State University of New York Health Science Center at Brooklyn (Howard Minkoff, MD, Joseph Feldman, DrPH, Sheldon Landesman, MD, Michael Augenbraun, MD, Jack DeHovitz, MD, MPH, Tracey Wilson, PhD, Susan Holman, RN, MS); Community Advisors (Jeannette Carter, Alexandra Harry). Washington DC Metropolitan Consortium: Georgetown University Medical Center (Mary Young, MD, Melanie Bacon, RN, MPH, MA, Qiang Yao, PhD); Howard University Medical Center (Robert Delapenha, MD); George Washington University Medical Center (Sylvia Silver, DA); Whitman-Walker Clinic (Kunthavi Sathasivam, MD); Montgomery County Health Department (Carol Jordan, RN, MPH); Inova Health System of Northern Virginia (Peggy Beckman, RN, CANP); Community Advisors (Kimberley Kelsey, Theresa Holt). The Connie Wofsy Study Consortium of Northern California: University of California, San Francisco (Ruth Greenblatt, MD, Herminia Palacio, MD, MPH, Nancy Hessol, MSPH, Nancy Padian, PhD, Deborah Greenspan, BDS, Meg Newman, MD); Alameda County Medical Center (Kathleen Clanon, MD); Alta Bates Hospital (Claire Borkert, MD); Community Advisors (Donna Haggerty, Nilda Rodriguez). Los Angeles County/Southern California Consortium: University of Southern California Medical Center/Los Angeles County (Alexandra Levine, MD, Andrea Kovacs, MD, Jean Richardson, PhD, Yvonne Barranday, BS); Charles Drew University of Medicine/Martin Luther King Medical Center (Wilber Jordan, MD); AIDS Healthcare Foundation (Charles Farthing, MD); the Santa Barbara County Department of Health Services (Elliott Schulman, MD); T.H.E. Clinic for Women (William Merritt, MD), Prototypes/WARN (Vivian Brown, PhD); University of Hawaii (Cecilia Shikuma, MD); Community Advisor (Elisa Sanchez-Jenkins). Chicago Consortium: Cook County Hospital (Mardge Cohen, MD, Audrey French, MD, Kathleen Weber, BSN); University of Illinois at Chicago (Ronald Hershow, MD); Rush Presbyterian–St. Luke's Medical Center (Beverly Sha, MD, L. Stewart Massad, MD, James Bremer, PhD, Alan Landay, PhD); Northwestern Memorial Hospital (Patricia Garcia, MD, MPH); Community Advisors (Marta Santiago, Chinniese Peterson). Data Coordinating Center: The Johns Hopkins School of Hygiene and Public Health, Baltimore, MD (Alvaro Muñoz, PhD, Stephen J. Gange, PhD, Linda Ahdieh, PhD, Jean Anderson, MD, Lisa P. Jacobson, ScD, Lynn Kirstein, MS, Cynthia Kleeberger, MAS, Joseph B. Margolick, MD, PhD, Sol Su, ScD, Patrick Tarwater, PhD). NIH: National Institute of Allergy and Infectious Diseases (Carolyn Williams, PhD); National Institute of Child Health and Human Development (Heather Watts, MD); National Institute of Dental and Craniofacial Research (Maryann Redford, DDS, MPH); National Institute of Drug Abuse (Katherine Davenny, MPH, Vincent Smeriglio, PhD); National Cancer Institute (Sandra L. Melnick, DrPH, Vaurice Starks, BS); Agency for Health Care Policy and Research, Centers for Disease Control and Prevention (Dawn Smith, MD, Dora Warren, PhD). Cited Here...